Data gathering is a fine art that many people haven’t mastered. In business and in science, you will often see sets of data that have been gathered using suboptimal techniques, leading to poor quality predictions and marring any insight that you might gain from analyzsing these data sets.
In general, for most purposes, you want to gather quantitative data. Quantitative data is a data type which revolves around collecting numerical data rather than qualitative data. Qualitative data is much more ephemeral, lacking the impact and being much less useful than quantitative data. You will find that you can act upon quantitative data, as the numbers are easily measurable. Qualitative data is often collected in fields where you cannot gain numerical answers, such as psychology and other social sciences. If you are having trouble devising a system that allows you to collect sufficient quality and quantity of qualitative data for your business or project, you could try out this Inferential Statistics in SPSS course, which is based in the SPSS statistical analysis program written by IBM, but can be applied to any type of statistics work.
What Exactly are Quantitative Techniques?
The term quantitative techniques covers a broad range of statistical gathering techniques that are all focused on getting numerical data for statistical analysis. These statistics are often then used for research and analysis leading to business decisions.
An example of a quantitative data point might be that “57% of our customers in the city of Chicago preferred the new formula fizzy drink to the old formula fizzy drink.”
This would be a good piece of data to know, as the company in question may decide to swap over to their new drink, in order to increase customer perceptions of the brand.
These quantitative results can come from a wide variety of sources. The best quality data that you can get, in terms of quantitative techniques, is the double blind test.
A double blind test gives the most accurate results as any bias that might occur in the test subjects or the tester will not be represented in the result. A double blind test makes it so both the tester and the test subjects in the experiment do not know the true reason for the experiment and are often told that the experiment is testing something else completely. Whilst you could just ask the test subjects to be open minded and try not to use bias over the course of the experiment, it is often unconscious bias that is reflected in the data sets. If you are testing a drug that makes people stronger, you give half of the test subjects a fake, sugar pill which is often known as a placebo, whilst giving the other half of the group the real drug. If you were to give everyone the same drug, any effects you see in the patients may not be because of the drug as the human mind is very easily manipulated. The placebo group are used as a type of baseline for any experiment.
In business, there are many other types of quantitative techniques you might apply to your data.
All quantitative techniques fall broadly under the umbrellas of mathematical, statistical, or programming based techniques and each has their own benefits and drawbacks. Most businesses will use multiple techniques simultaneously as this will give the company a more rounded picture of how to use the data correctly. Quantitative techniques are much more accurate than Qualitative techniques, as they eliminate the bias associated with both qualitative tests and non blind tests.
A popular type of quantitative technique is differentiation. Differentiation is a mathematical process involving calculus and it is useful for seeing change over time within a given system. Differentiation is generally used to figure out the changes in a system when a variable in the system changes, measuring how the end result changes by altering a variable. This could be used in many ways: in cooking, chemistry, and many physical sciences, yet it is less useful in a social science. Differentiation also has an opposite, integration, which works in the opposite way. Integration is used to see the changes to a variable when the system changes.
Both of these are valuable quantitative techniques to learn and are very difficult to get your head around. The mathematics involved is very high level and people often struggle with it even after being taught how to do it. This course in Integral Calculus will serve you well in remembering or even learning calculus for the first time. The course is fantastic and will help you develop your quantitative data analysis techniques and also will teach you in easy understandable steps how to use calculus for many different situations.
Regression analysis is incredibly useful and a whole host of people use this technique every single day in their business life. Generally, economists are interested in the concept of regression analysis, which is based around finding a causal link or correlation between two independent variables in any given system. A common example for regression analysis is that of measuring the salary of an employee and their level of education, to see if there is a correlation between the two factors. You could also use this in cooking and many other fields, as you can see. Regression analysis is useable in many fields and will save you time if you learn how to use it and integrate in to your business.
Regression analysis uses two sets of data, predictors and independent variables. These values can be anything, from total revenue to tax rate to advertisement budgets and so on. Comparing the two is the basis of regression analysis.
Simulation is a great way to get pseudo real world data on anything that can be simulated effectively in a controlled environment. If you can simulate a scenario effectively, you can then see how test subjects respond to stressors and often this information is very valuable. It’s not just used for living things however, a wind tunnel is a widely used simulator to test the aerodynamics of cars and other objects. This data allows the manufacturer to make tweaks in design and concept and can show data which may lead to the product being discontinued before production starts. This is obviously a good thing as recalling product lines is costly and should be avoided at all costs. Simulation allows these kind of usability tests, even in unlikely scenarios.
Factor analysis is another often used data technique used for quantitative data analysis. This type of analysis tries to thin down the amount of data that is available to be used by exploring the similarities between multiple sets of data. This way, you can analyze the overall trends that are hiding in the data without having to figure these out yourself. Market researchers and economists are very avid users of Factor Analysis as it makes trawling through large sets of data received from surveys easy and quick.
Another one for the economists, indexes are a fantastic way to use quantitative research to simplify and share data with the general public in an efficient and easy manner. Indexes are all over the finance world, with each of the major stock exchanges (NASDAQ etc.) having an index as a representation of how the financial market is doing. Analyzing indexes is useful as they are a useful way to see how the overall trend of a given environment is behaving. People base decisions worth hundreds of thousands of dollars on the existence of stock market indexes every day and without quantitative analysis and research this wouldn’t be possible.
Game and Probability Theory
Game Theory is a class of thought that aims to find the most optimal strategy in any given scenario. It achieves this by using quantitative methods and thought experiments and always finds the optimal course of action in a competitive situation. This type of quantitative technique is slightly less applicable to business, yet very useful if you find yourself in a situation where you are unsure of the options.
The “Prisoner’s Dilemma” is a very common instance of game theory, showing why two people may not cooperate with each other, even if cooperation is the best move, statistically. In the canon of the Prisoner’s Dilemma, two prisoners are offered one of three options to remedy their current situation. They are offered the chance to testify about the other person, getting released from prison at the expense of the other prisoner spending times in the jail. The other option is to be quiet, not telling the officer anything. If both parties stay silent, both parties are in jail for a year, whereas in the other scenario, the jail time served is greater. This shows that the correct thing to do is stay silent. Of course, in that situation, you would probably try to get out of jail by offering evidence against the others. The dilemma occurs as both parties are given these options, meaning if both parties try to get no jail time, they both end up in jail for longer. Quantitative thinking techniques like these allow people to make more logical and useful real world decisions and are a cornerstone of advanced logical reasoning.
Probability theory is useful to use in conjunction with statistics allowing someone to semi-accurately predict how someone or something will act in a given situation, assuming you have access to all of the necessary data. Probability theory is useable to see patterns in apparent randomness. This is how we know that the probability of getting a heads or tails on any given coin flip is equal. The coin flip itself is random, yet over time it averages out to a 50% chance of heads or tails.
Quantitative Data Collection
As mentioned above, the best way to collect non-biased and useful quantitative data is choosing to conduct double or triple blind experiments which allow more accurate results for a given portion size. Quantitative data can also be collected in many other ways, depending on the situation you are trying to gather data upon. For data from inanimate objects, you can use sensors and electronic surveying tools to gather numerical data. When you are trying to get quantitative data from people, it is a little more difficult to get accurate data. Surveys and questionnaires will get you some useable data but the data from these may be inaccurate. Many people will answer untruthfully on a questionnaire for lots of different reasons. If you need to find data on objects or the general population, city records and other standardized records will be of a great help to you. For products, most manufacturers will keep records of their product specifications, for the general public to browse. This is helpful for techniques, which need complete information to use, such as testing audio or visual equipment and tests of this sort.
Whilst quantitative data is useful for many things, you will find that it will not give you all the answers. Qualitative data is still useful as it allows you to gain an understanding of the motivations of the subject and can be used to figure out things about prevalent thought patterns.
Quantitative research is used more and more every day in almost all fields of commerce and science. If you feel you need to include some quantitative techniques in your work or would like to learn more about them, there are a huge number of resources on the web that can help you learn more about these techniques. Quantitative techniques allow you access to a huge amount of data from extrapolation that you wouldn’t be able to predict or use if there was no way to use quantitative techniques. Quantitative Research is a fantastic course that will explain more about these techniques and how you can integrate them in to your businesses and day to day workflow.